Bypass Attacks That Ignore the Lock Entirely

There is a category of entry that makes lock ratings irrelevant, and it is worth understanding precisely because it is invisible on a spec sheet. A bypass does not pick, drill, or force the cylinder. It reaches around it to the thing the cylinder was supposed to be guarding: the latch, the thumbturn, or the opening itself.

The mental model that helps here is that a lock has two separate jobs. It decides who is authorized, and it physically holds the door shut. Cylinder security addresses the first job. Bypass attacks the second, or skips both by using a gap that has nothing to do with the lock at all.

The three broad categories

Described at the level of what they exploit rather than how they are performed:

• Latch manipulation. A spring latch with a sloped face is designed to retract when pushed. If the latch is reachable from the exterior side of the door, that design behavior can be triggered without a key. This is a property of an unsecured spring latch, not a defect in a particular brand.
• Reaching an interior control. Thumbturns, interior levers, and panic hardware are all designed to open the door easily from inside. If there is any aperture through which the interior side can be reached, that ease of operation now works for whoever can reach it. Glass panels adjacent to the lock, mail slots, and pet doors are the recurring examples.
• The gap itself. Undercuts, worn weatherstripping, poorly fitted frames, and oversized clearances create paths that never involve the lock. Some of these are byproducts of ventilation or ADA thresholds and cannot simply be sealed, which is why the answer is usually to move the vulnerable control rather than to close the gap.

Why deadlatches exist

Hardware engineers addressed latch manipulation decades ago with the deadlatch, and the patent record documents the mechanism clearly. A deadlatch adds a small auxiliary plunger alongside the main latch bolt. When the door closes properly, the plunger is held depressed by the strike plate, and that depressed position engages a blocking element that prevents the main latch from retracting under side pressure. A representative example is US Patent 5,100,184 for a deadlatch assembly, which describes a cam member whose bosses seat into recesses to physically prevent the latch bolt from sliding to the unlocked position.

Two practical consequences follow. First, a deadlatch only works if it actually lands on the strike plate rather than falling into the strike opening, which makes strike alignment a security issue and not merely a fit and finish issue. Second, a properly thrown deadbolt has no sloped face and no spring, so it is not susceptible to this category at all. That is the real argument for a deadbolt over a latch alone.

Where standards address this

Manipulation resistance is a tested property, not a marketing term. ANSI/BHMA A156.36 covers auxiliary locks, which is the category deadbolts fall into, and its published scope includes five classifications of tests: operational, cycle, strength, security, and finish, with products graded from 1 to 3. For cylinders specifically, ANSI/BHMA A156.30 sets security performance based requirements for high security cylinders covering the cylinder, its detainers, and its tailpiece or cam.

Note what that scope does and does not cover. It covers the cylinder assembly. It does not cover your sidelight, your mail slot, or the alignment of your strike. Certification is a statement about a component under laboratory conditions, and both BHMA standards say so directly.

A defensive inspection of your own openings

Walk each exterior door and ask:

• Is there glass within arm's reach of any interior thumbturn or lever? If yes, that is your highest priority finding.
• Does the door have a mail slot or pet door? Where does the opening sit relative to the interior hardware?
• When the door is closed, does the auxiliary plunger rest on the strike plate face, or has misalignment let it drop into the hole?
• Does the deadbolt throw fully and freely, with no need to lift or push the door?
• Is there a deadbolt at all, or is a spring latch in a knob doing the entire job?
• Are the gaps between door and frame consistent, or has the door or frame moved?

Most findings from that list are inexpensive to correct. Strike alignment, adding a deadbolt, and relocating or shielding hardware away from glazing address the majority of bypass exposure without touching the cylinder. If a fix would affect egress, such as anything involving how quickly occupants can get out, talk to a locksmith or your local building authority first. Life safety codes constrain what is permitted, and they exist for good reasons.